Mn_fit can perform a wide variety of mathematical fits and analyses on data or histograms read in from a file. the program is interactive with a graphical display, so it is easy for the user to explore different fits while viewing the results. many functions are built into the package, and new ones can be added.
some capabilities of mn_fit include built-in maximum likelihood fits, fitting of 1-d and 2-d histograms, fitting several histograms simultaneously, applying constraints on function parameters, and limited use of monte carlo statistics. the program can produce postscript output of a quality suitable for publication. mn_fit is based on time-tested cernlib software including minuit, comis, and higz.
to build a mn_fit executable that includes custom user-defined functions, you also must install the mn-fit-dev package.